Shunyu Liu (刘顺宇)


(Last Updated: Sep 26, 2024)

I am currently a research scientist at Nanyang Technological University, working with Prof. Dacheng Tao. Before that, I received the Ph.D. degree from the College of Computer Science and Technology at Zhejiang University in 2024, advised by Prof. Mingli Song and Prof. Chun Chen, and received the B.Eng. degree in Software Engineering from Sun Yat-sen University in 2019.

My research interests include multi-agent learning, reinforcement learning and LLM agent. Applications of my work include autonomous power system control, as well as applications in other decision-making domains. The long-term goal of my research is to develop efficient, generalized, and practical agents. In tandem with this, my research strives to facilitate intelligent interaction among multiple agents, empowering them to tackle complex decision-making challenges in both the virtual and real worlds.

Please feel free to contact me if you are interested in my research :)


News



Selected Publications


$^\dagger$ denotes equal contribution, and $^✉️$ denotes the corresponding author.

  • Odyssey: Empowering Minecraft Agents with Open-World Skills
    Shunyu Liu$^\dagger$, Yaoru Li$^\dagger$, Kongcheng Zhang$^\dagger$, Zhenyu Cui$^\dagger$, Wenkai Fang$^\dagger$, Yuxuan Zheng, Tongya Zheng, Mingli Song$^✉️$.
    arXiv preprint arXiv:2407.15325, 2024.
    [arXiv] [Code]

  • Interaction Pattern Disentangling for Multi-Agent Reinforcement Learning
    Shunyu Liu, Jie Song$^✉️$, Yihe Zhou, Na Yu, Kaixuan Chen, Zunlei Feng, Mingli Song.
    IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)$^*$, 2024.
    $^*$ Top-tier Journal in Artificial Intelligence.
    [Paper] [arXiv] [Code]

  • Transmission Interface Power Flow Adjustment: A Deep Reinforcement Learning Approach based on Multi-task Attribution Map
    Shunyu Liu$^\dagger$, Wei Luo$^\dagger$, Yanzhen Zhou, Kaixuan Chen, Quan Zhang, Huating Xu, Qinglai Guo, Mingli Song$^✉️$.
    IEEE Transactions on Power Systems (TPS)$^*$, 2024.
    $^*$ Top-tier Journal in Power Systems.
    [Paper] [arXiv] [Code]

  • Ask-AC: An Initiative Advisor-in-the-Loop Actor-Critic Framework
    Shunyu Liu, Kaixuan Chen, Yu, Na, Jie Song, Zunlei Feng, Mingli Song$^✉️$.
    IEEE Transactions on Systems, Man and Cybernetics: Systems (TSMC), 2023.
    [Paper] [arXiv] [Code]

  • Contrastive Identity-Aware Learning for Multi-Agent Value Decomposition
    Shunyu Liu$^\dagger$, Yihe Zhou$^\dagger$, Jie Song$^✉️$, Tongya Zheng, Kaixuan Chen, Tongtian Zhu, Zunlei Feng, Mingli Song.
    AAAI Conference on Artificial Intelligence (AAAI), 2023, Oral.
    [Paper] [arXiv] [Code]


Awards & Honors


Scholarship

  • Outstanding Undergraduate Student Scholarship, Sun Yat-sen University, 2016, 2017, 2018.
  • National Scholarship, Ministry of Education of the People’s Republic of China, 2018.
  • Outstanding Graduate Student Award, Zhejiang University, 2020, 2021, 2023.
  • Outstanding Ph.D. Student Scholarship, Zhejiang University, 2021.
  • Chen Tianzhou Scholarship, Zhejiang University, 2023.

Competition

  • Meritorious Winner in “Mathematical Contest in Modeling”, COMAP, 2018.
  • Ranked 9th in “Advanced Power System AI Application Competition”, State Grid Corporation of China, 2022
  • Ranked 5th in “L2RPN Challenge - Energies of the Future and Carbon Neutrality”, WCCI, 2022.
  • Ranked 3rd in “Power System Dispatching AI Application Competition”, China Southern Power Grid, 2022.
  • Ranked 3rd in “Power System Dispatching AI Application Competition”, China Southern Power Grid, 2023.


Academic Services


Journal Reviewer

  • IEEE Transactions on Knowledge and Data Engineering (TKDE)
  • Information Sciences (INS)
  • Autonomous Agents and Multi-Agent Systems (AAMAS)

Conference Reviewer

  • NeurIPS 2024, KDD 2024, ICLR 2023, ECAI 2023, ICONIP 2023